Constraints in estimating the proton density fat fraction

Bydder, Mark; Ghodrati, Vahid; Gao, Yu; Robson, Matthew D; Yang, Yingli; Hu, Peng

doi:10.1016/j.mri.2019.11.009

Cited by 16 publications

(20 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…No additional preprocessing was necessary for the T1w 3D data for the pancreas. Proton density fat fraction (PDFF) and R2* were estimated from the single-slice multi-echo data for the liver and pancreas ( Bydder et al, 2020b ). The R2* values were converted into iron concentrations ( McKay et al, 2018 ; Wood et al, 2005 ).…”

Section: Methodsmentioning

confidence: 99%

Genetic architecture of 11 organ traits derived from abdominal MRI using deep learning

Liu

Basty

Whitcher

et al. 2021

eLife

138

124

View full text Add to dashboard Cite

Cardiometabolic diseases are an increasing global health burden. While socioeconomic, environmental, behavioural, and genetic risk factors have been identified, a better understanding of the underlying mechanisms is required to develop more effective interventions. Magnetic resonance imaging (MRI) has been used to assess organ health, but biobank-scale studies are still in their infancy. Using over 38,000 abdominal MRI scans in the UK Biobank, we used deep learning to quantify volume, fat, and iron in seven organs and tissues, and demonstrate that imaging-derived phenotypes reflect health status. We show that these traits have a substantial heritable component (8–44%) and identify 93 independent genome-wide significant associations, including four associations with liver traits that have not previously been reported. Our work demonstrates the tractability of deep learning to systematically quantify health parameters from high-throughput MRI across a range of organs and tissues, and use the largest-ever study of its kind to generate new insights into the genetic architecture of these traits.

show abstract

Section: Methodsmentioning

confidence: 99%

Genetic architecture of 11 organ traits derived from abdominal MRI using deep learning

Liu

Basty

Whitcher

et al. 2021

eLife

138

124

View full text Add to dashboard Cite

show abstract

“…systems with phantoms(48, 82,(112)(113)(114). For the MRIdian 0.35T MRI, a few DWI studies have been performed, demonstrating the ADC accuracy and reproducibility, as well as improving DWI spatial integrity(48,112).…”

mentioning

confidence: 99%

“…For the MRIdian 0.35T MRI, a few DWI studies have been performed, demonstrating the ADC accuracy and reproducibility, as well as improving DWI spatial integrity(48,112). Studies of Nejad-Davarini et al (82) and Bydder et al(113) also explored feasibility and accuracy of T1 mapping, R2* mapping, proton density mapping, and proton density fat fraction using MRIdian. A multicenter study showed that consistent ADC, T1, T2, and DCE values can be measured across institutes with a Unity system…”

mentioning

confidence: 99%

Quantitative Magnetic Resonance Imaging for Biological Image-Guided Adaptive Radiotherapy

2021

Self Cite

View full text Add to dashboard Cite

MRI-guided radiotherapy systems have the potential to bring two important concepts in modern radiotherapy together: adaptive radiotherapy and biological targeting. Based on frequent anatomical and functional imaging, monitoring the changes that occur in volume, shape as well as biological characteristics, a treatment plan can be updated regularly to accommodate the observed treatment response. For this purpose, quantitative imaging biomarkers need to be identified that show changes early during treatment and predict treatment outcome. This review provides an overview of the current evidence on quantitative MRI measurements during radiotherapy and their potential as an imaging biomarker on MRI-guided radiotherapy systems.

show abstract

“…This acquisition was stopped for the liver after the first 10,000 subjects (approximately) and replaced by the IDEAL sequence, but was continued for the pancreas for all subjects. A single-slice IDEAL sequence (Reeder et al, 2005) for the liver used the following parameters: TR = 14 ms, TE = 1.2/3.2/5.2/7.2/9.2/11.2 ms, FA = 5°, bandwidth = 1565 Hz, voxel size 1.719 x 1.719 x 10.0 mm and 256 x 232 matrix. We applied bias-field correction to each echo time separately to facilitate 2D segmentation. Software (https://github.com/marcsous/pdff) available from Dr Mark Bydder, specifically the PRESCO (Phase Regularized Estimation using Smoothing and Constrained Optimization) algorithm (Bydder et al, 2020), was used to simultaneously estimate the proton density fat fraction (PDFF, referred to as fat in results) and transverse relaxivity (R2*) values voxelwise from the single-slice gradient echo (GRE) and IDEAL acquisitions. Essentially, a multi-peak spectrum was constructed from the echo times in the acquisition protocol and used to perform nonlinear least squares under multiple regularization constraints that extends the IDEAL (Iterative Decomposition of Water and Fat with Echo Asymmetry and Least-Squares Estimation) algorithm (Reeder et al, 2005; Yu et al, 2008).…”

Section: Methodsmentioning

confidence: 99%

“…The single-slice data (Gradient Multi-Echo (GRE) and IDEAL) used for liver and pancreas quantification had bias-field correction applied to each echo time separately for 2D segmentation. We used the PRESCO (Phase Regularized Estimation using Smoothing and Constrained Optimization) algorithm 75 to simultaneously estimate PDFF (referred to as fat in results) and R2* values voxelwise from the (uncorrected) GRE and IDEAL acquisitions. For consistency with previous studies 55, 76 , we convert R2* into iron concentration (mg/g) using the formula: iron concentration = 0.202 + 0.0254 x R2*.…”

Section: Image Preprocessing Pipelinementioning

confidence: 99%

Genetic architecture of 11 abdominal organ traits derived from abdominal MRI using deep learning

Liu

Basty

Whitcher

et al. 2020

Preprint

View full text Add to dashboard Cite

Cardiometabolic diseases are an increasing population health burden and while several well established socioeconomic, environmental, behavioural, and genetic risk factors have been identified, our understanding of their drivers and mechanisms remains incomplete. Thus, a better understanding of these factors is required for the development of more effective interventions. Magnetic resonance imaging (MRI) has been used to assess organ health in a number of studies, but large-scale population-based studies are still in their infancy. Using deep learning to segment individual organs from up to 38,683 abdominal MRI scans in the UK Biobank, we demonstrate that image derived phenotypes such as volume, fat and iron content reflect overall organ health. We further show that these traits have a substantial heritable component which is enriched in organ-specific cell types. We also identify several novel genome-wide significant associations. Overall our work demonstrates the feasibility and power of high-throughput MRI for the multi-organ study of cardiometabolic disease, health, and ageing.

show abstract

Constraints in estimating the proton density fat fraction

Cited by 16 publications

References 30 publications

Genetic architecture of 11 organ traits derived from abdominal MRI using deep learning

Genetic architecture of 11 organ traits derived from abdominal MRI using deep learning

Quantitative Magnetic Resonance Imaging for Biological Image-Guided Adaptive Radiotherapy

Genetic architecture of 11 abdominal organ traits derived from abdominal MRI using deep learning

Contact Info

Product

Resources

About